在计算机视觉领域,由镜头切换、目标动力学突变、低帧率视频等引起的突变运动存在极大的不确定性,使得突变运动跟踪成为该领域的挑战性课题.以贝叶斯滤波框架为基础,提出一种基于有序超松弛Hamiltonian马氏链蒙特卡罗方法的突变运动跟...在计算机视觉领域,由镜头切换、目标动力学突变、低帧率视频等引起的突变运动存在极大的不确定性,使得突变运动跟踪成为该领域的挑战性课题.以贝叶斯滤波框架为基础,提出一种基于有序超松弛Hamiltonian马氏链蒙特卡罗方法的突变运动跟踪算法.该算法将Hamiltonian动力学融入MCMC(Markov chain Monte Carlo)算法,目标状态被扩张为原始目标状态变量与一个动量项的组合.在提议阶段,为抑制由Gibbs采样带来的随机游动行为,提出采用有序超松弛迭代方法来抽取目标动量项.同时,提出自适应步长的Hamiltonian动力学实现方法,在跟踪过程中自适应地调整步长,以减少模拟误差.提出的跟踪算法可以避免传统的基于随机游动的MCMC跟踪算法所存在的局部最优问题,提高了跟踪的准确性而不需要额外的计算时间.实验结果表明,该算法在处理多种类型的突变运动时表现出出色的处理能力.展开更多
针对运动突变目标视觉跟踪问题,提出一种基于视觉显著性的两阶段采样跟踪算法.首先,将视觉显著性信息引入到Wang-Landau蒙特卡罗(Wang-Landau Monte Carlo,WLMC)跟踪算法中,设计了结合显著性先验的接受函数,利用子区域的显著性值来引导...针对运动突变目标视觉跟踪问题,提出一种基于视觉显著性的两阶段采样跟踪算法.首先,将视觉显著性信息引入到Wang-Landau蒙特卡罗(Wang-Landau Monte Carlo,WLMC)跟踪算法中,设计了结合显著性先验的接受函数,利用子区域的显著性值来引导马尔可夫链的构造,通过增大目标出现区粒子的接受概率,提高采样效率;其次,针对运动序列中平滑与突变运动共存的特点,建立两阶段采样模型.其中第一阶段对目标当前运动类型进行判定,第二阶段则根据判定结果采用相应算法.突变运动采用基于视觉显著性的WLMC算法,平滑运动采用双链马尔可夫链蒙特卡罗(Marko chain Monte Carlo,MCMC)算法,以此完成目标跟踪,提高算法的鲁棒性.该算法既避免了目标在平滑运动时全局采样导致精度下降的缺点,又能在目标发生运动突变时有效捕获目标.实验结果表明,该算法不仅能有效处理运动突变目标的跟踪问题,在典型图像序列上也具有良好的鲁棒性.展开更多
Object tracking with abrupt motion is an important research topic and has attracted wide attention.To obtain accurate tracking results,an improved particle filter tracking algorithm based on sparse representation and ...Object tracking with abrupt motion is an important research topic and has attracted wide attention.To obtain accurate tracking results,an improved particle filter tracking algorithm based on sparse representation and nonlinear resampling is proposed in this paper. First,the sparse representation is used to compute particle weights by considering the fact that the weights are sparse when the object moves abruptly,so the potential object region can be predicted more precisely. Then,a nonlinear resampling process is proposed by utilizing the nonlinear sorting strategy,which can solve the problem of particle diversity impoverishment caused by traditional resampling methods. Experimental results based on videos containing objects with various abrupt motions have demonstrated the effectiveness of the proposed algorithm.展开更多
文摘在计算机视觉领域,由镜头切换、目标动力学突变、低帧率视频等引起的突变运动存在极大的不确定性,使得突变运动跟踪成为该领域的挑战性课题.以贝叶斯滤波框架为基础,提出一种基于有序超松弛Hamiltonian马氏链蒙特卡罗方法的突变运动跟踪算法.该算法将Hamiltonian动力学融入MCMC(Markov chain Monte Carlo)算法,目标状态被扩张为原始目标状态变量与一个动量项的组合.在提议阶段,为抑制由Gibbs采样带来的随机游动行为,提出采用有序超松弛迭代方法来抽取目标动量项.同时,提出自适应步长的Hamiltonian动力学实现方法,在跟踪过程中自适应地调整步长,以减少模拟误差.提出的跟踪算法可以避免传统的基于随机游动的MCMC跟踪算法所存在的局部最优问题,提高了跟踪的准确性而不需要额外的计算时间.实验结果表明,该算法在处理多种类型的突变运动时表现出出色的处理能力.
文摘针对运动突变目标视觉跟踪问题,提出一种基于视觉显著性的两阶段采样跟踪算法.首先,将视觉显著性信息引入到Wang-Landau蒙特卡罗(Wang-Landau Monte Carlo,WLMC)跟踪算法中,设计了结合显著性先验的接受函数,利用子区域的显著性值来引导马尔可夫链的构造,通过增大目标出现区粒子的接受概率,提高采样效率;其次,针对运动序列中平滑与突变运动共存的特点,建立两阶段采样模型.其中第一阶段对目标当前运动类型进行判定,第二阶段则根据判定结果采用相应算法.突变运动采用基于视觉显著性的WLMC算法,平滑运动采用双链马尔可夫链蒙特卡罗(Marko chain Monte Carlo,MCMC)算法,以此完成目标跟踪,提高算法的鲁棒性.该算法既避免了目标在平滑运动时全局采样导致精度下降的缺点,又能在目标发生运动突变时有效捕获目标.实验结果表明,该算法不仅能有效处理运动突变目标的跟踪问题,在典型图像序列上也具有良好的鲁棒性.
基金Supported by the National Natural Science Foundation of China(61701029)
文摘Object tracking with abrupt motion is an important research topic and has attracted wide attention.To obtain accurate tracking results,an improved particle filter tracking algorithm based on sparse representation and nonlinear resampling is proposed in this paper. First,the sparse representation is used to compute particle weights by considering the fact that the weights are sparse when the object moves abruptly,so the potential object region can be predicted more precisely. Then,a nonlinear resampling process is proposed by utilizing the nonlinear sorting strategy,which can solve the problem of particle diversity impoverishment caused by traditional resampling methods. Experimental results based on videos containing objects with various abrupt motions have demonstrated the effectiveness of the proposed algorithm.
